NASA/TM-2011-216302

.

~N~~/. ~".,

Low Force Icy Regolith PenetrationTechnology

Philip T. Metzger, Gregory M. Galloway, and James G. MantovaniNASA Kennedy Space Center

Kris Zacny and Jack CraftHoneybee Robotics, Inc., New York City, New York

July 11, 2011

The NASA STI Program Office .. .in Profile

Since its founding, NASA has been dedicated tothe advancement of aeronautics and spacescience. The NASA Scientific and TechnicalInformation (STI) Program Office plays a keypart in helping NASA maintain this importantrole.

The NASA STI Program Office is operated byLangley Research Center, the lead center forNASA's scientific and technical information.The NASA STI Program Office provides accessto the NASA STI Database, the largestcollection of aeronautical and space science STIin the world. The Program Office is alsoNASA's institutional mechanism fordisseminating the results of its research anddevelopment activities. These results arepublished by NASA in the NASA STI ReportSeries, which includes the following reporttypes:

• TECHNICAL PUBLICATION. Reports ofcompleted research or a major significantphase of research that present the results ofNASA programs and include extensive dataor theoretical analysis. Includescompilations of significant scientific andtechnical data and information deemed to beof continuing reference value. NASAcounterpart of peer-reviewed formalprofessional papers, but having less stringentlimitations on manuscript length and extentof graphic presentations.

• TECHNICAL MEMORANDUM.Scientific and technical findings that arepreliminary or of specialized interest, e.g.,quick release reports, working papers, andbibliographies that contain minimalannotations. Does not contain extensiveanalysis.

• CONTRACTOR REPORT. Scientific andtechnical findings by NASA-sponsoredcontractors and grantees.

• CONFERENCE PUBLICATION.Collected papers from scientific andtechnical conferences, symposia, seminars,or other meetings sponsored or co-sponsoredby NASA.

• SPECIAL PUBLICATIONS. Scientific,technical, or historical information fromNASA programs, projects, and missions,often concerned with subjects havingsubstantial public interest.

• TECHNICAL TRANSLATION. English­language translations of foreign scientificand technical material pertinent to NASA'smission.

Specialized services to complement the STIProgram Office's diverse offerings includecreating custom thesauri, building customizeddatabases, organizing and publishing researchresults ... even providing videos.

For more information about the NASA STIProgram Office, see the following:

• Access the NASA STI Program Home Pageat http://www.sti.nasa.gov

• E-mail your question via the internet tohelp@sti.nasa.gov

• Fax your question to the NASA STI HelpDesk at (301) 621-0134

• Phone the NASA STI Help Desk at (301)621-0390

• Write to:NASA STI Help DeskNASA Center for Aerospace Information1721 Standard DriveHanover, MD 21076-1320

NASA/TM-2011-216302

. .

..~. .

Low Force Icy Regolith PenetrationTechnology

Philip T. Metzger, Greg M. Galloway, and James G. MantovaniNASA Kennedy Space Center

Kris Zacny and Jack CraftHoneybee Robotics Space Mechanisms CorporationNew York, New York

National Aeronautics andSpace Administration

Kennedy Space Center

July 11, 2011

Acknowledgments

This work was supported by the 2010 NASA Innovation Fund from NASA's InnovativePartnerships Program and by cost sharing by Honeybee Robotics Space MechanismsCorporation. The NASA contract with Honeybee for this project was NNK10EA98P. Honeybeeprovided to NASA a final contract report HBR Doc No. 249.RPT.001 dated September 2,2010.Portions of that Honeybee report have been incorporated in this document with permission of theauthors.

Available from:

NASA Center for AeroSpace Information (CAS!)7121 Standard DriveHanover, MD 21076-1320(301) 621-0390

National Technical Information Service (NTIS)5285 Port Royal RoadSpringfield, VA 22161-2171(703) 605-6000

This report is also available in electronic form at URL http://www.sti.nasa.gov/ andhttp://ntrs.nasa.gov/search.jsp.

11

1-------------------------------------Contents

ACKNOWLEDGMENTS II

FIGURES IV

TABLES V

PREFACE 7

INTRODUCTION 8

EXPERIMENTS 8

Percussive Cone Penetrometer 8

Ice/Soil Mixtures 9

Procedure 11

RESULTS 12

DISCUSSION 17

Ice Content 17

Penetration Rate 18

Retracting the Rod 19

SUMMARY OF FINDINGS 19

CONCLUSIONS 20

REFERENCES 21

111

Figures

Figure 1. Honeybee Robotics percussive cone penetrometer. 9

Figure 2. (Left) Ice/soil can. (Right) I-meter column 10

Figure 3. One-meter icy/soil column with final position of penetrometer. 11

Figure 4. Penetration into sample can 12

Figure 5. Penetration depth versus time for sample can #3 13

Figure 6. Downforce versus time for sample can #3 13

Figure 7. Penetration depth versus time for sample can #4 14

Figure 8. Downforce versus time for sample can #4 14

Figure 9. Fractured surface of can #4. Shaft demonstrates depth of penetration 15

Figure 10. Fractured surface of can #5 with 100% water ice 15

Figure 11. Depth of penetration and downforce versus time for I-meter column 16

Figure 12. Top surface of I-meter column with embedded penetrometer rod 17

IV

TablesTable 1: SoillIce Sample Cans 10

v

This page intentionally left blank.

VI

PrefaceRecent data from the Moon, including LCROSS data, indicate large quantities of water ice and other

volatiles frozen into the soil in the permanently shadowed craters near the poles. If verified andexploited, these volatiles will revolutionize spaceflight as an inexpensive source of propellants and otherconsumables outside Earth's gravity well. This report discusses a preliminary investigation of a methodto insert a sensor through such a soil/ice mixture to verify the presence, nature, and concentration of theice. It uses percussion to deliver mechanical energy into the frozen mixture, breaking up the ice anddecompacting the soil so that only low reaction forces are required from a rover or spacecraft to push thesensor downward. The tests demonstrate that this method may be ideal for a small platform in lunargravity. However, there are some cases where the system may not be able to penetrate the icy soil, andthere is some risk of the sensor becoming stuck so that it cannot be retracted, so further work is needed.A companion project (lSDS for Water Detection on the Lunar Surface) has performed preliminaryinvestigation of a dielectric/thermal sensor for use with this system.

7

Introduction

Data from remote sensing and from the LCROSS mission indicate the existence of vast quantities ofwater ice and other volatiles in the lunar regolith, frozen into the permanently shadowed craters near thepoles. This will be game-changing because of the magnitude and importance of such a resource outsideEarth's gravity well. It will be important to ground-truth these findings by putting an instrument into thelunar regolith to measure the ice content directly. This will be difficult to do with a small-class ormedium-class rover in the low lunar gravity because these platforms will not have adequate weight toprovide the necessary downforce. In the Apollo program, astronauts found it extremely difficult to pushtubes into the dense, frictional lunar soil in 1/6 G. It will be even worse as we take core samples, anchoronto, and mine asteroids and small moons like Phobos, or mine icy soil on the Moon. Low-forcepenetration systems will be mandatory in all these situations.

Prior work by Honeybee Robotics Spacecraft Mechanisms Corporation has shown that percussivecone penetrometers are capable of penetrating lunar regolith with only a small fraction of the force of anordinary penetrometer. This study asks the question whether percussion is a suitable method to insertinstruments into that regolith when it contains various quantities of water ice. We performed preliminaryexperiments to measure penetration resistance in lunar soil simulant with varying quantities of water ice.We also performed a demonstration of percussive penetration into a I-meter deep column of ice and soilmixtures in layers of varying proportions. One meter is the expected depth to reach the ice beneath thedesiccated upper layers of lunar soil. A companion study (lSDS for Water Detection on the LunarSurface) has performed preliminary investigation of a dielectric/thermal sensor that may be inserted intothe regolith by this percussive penetration met~ to positively identify lunar ice. This system combiningthe sensor with percussive penetration ha~otential to provide the first-ever ground truth of vastquantities of dense ice layers in the lunar regolith.

ExperimentsThe experiments were performed jointly by Honeybee and NASA at the Kennedy Space Center

(KSC). The percussive penetrator was provided and operated by Honeybee Robotics. The mixtures oflunar soil simulant and water ice were prepared by NASA/KSC. Descriptions of the hardware andexperiments are provided below.

Percussive Cone Penetrometer

Originally we had planned to use both percussion and gas pulsing in the cone penetrometer. Furtheranalysis of the icy soil mixtures indicated that gas pulsing was not an appropriate approach due to the lowpermeability of ice-impregnated lunar soil and due to its extremely high mechanical strength. Therefore,the gas pulsing approach was abandoned. For this effort, we used a percussive dynamic conepenetrometer originally developed under a separate Honeybee SBIR Phase 1 effort. In the configurationtested, this device delivers 2.6 Joules of percussive energy per blow at a frequency of approximately1500-1750 blows per minute. This device was originally designed as a geotechnical instrument: Bydriving a cone into soil using a known percussive energy and recording the rate of penetration, soilstrength can be derived. This device is shown in Figure 1.

8

Figure 1. Honeybee Robotics percussive cone penetrometer.

Ice/Soil Mixtures

Lunar soil simulant JSC-1 A was used in this project to represent lunar soil. Carrier et al [1991] havesummarized the geotechnical properties of actual lunar soil without ice. The mechanics of JSC-1Awithout ice have been studied by Alshibli and Hasan [2009] and Zeng et al [2010]. The mechanics of theoriginal version of this simulant, JSC-1, was reported by Klosky et al [2000]. No data have been returnedfrom the Moon to indicate the mechanics of the soil with or without ice as it may be found in thepermanently shadowed craters. Gertsch et al [2006] used JSC-1A in ice/simulant mixtures toexperimentally study the resistance to a surface indentor (chipping the surface) and found that resistanceis a strong function of ice content. They reported that ice concentrations of 0.6 to 1.5% by mass behavelike weak shale or mudstone, whereas concentrations of 10.6% behave like strong limestone or sandstoneand thus would be very difficult to excavate. Gamsky and Metzger [2010] used JSC-I A without ice onshake tables and in ovens and report that iceless regolith in the permanently shadowed craters may be lesscompacted than elsewhere on the Moon due to the lack of the strong, localized, diurnal quakes to shakedown the soil and due to the lack of thermal cycling to directly compact it [Chen et ai, 2006]. We areunaware of any other studies addressing soil mechanics in the permanently shadowed craters. The JSC­1A in this study was dried thoroughly in an oven and massed then monitored as it cooled and re-adsorbedhumidity in the laboratory environment. The percent mass of adsorbed water was not significant, so pre­drying was not performed for further sample preparation. JSC-IA was mixed with water in percentagesranging between 0% and 8% by mass in 1% increments. The water/simulant mixtures were placed inlayers into six I-gallon cans (paint cans) as shown in Fig. 2 (left). Each can had three layers, the layerwith the greatest moisture content at the bottom. Each layer was tamped to compact it as it was laiddown. The moisture did not migrate significantly from their original layers to the adjacent layers due tothe relative impermeability of JSC-1A and because they were quickly frozen. Freezing was performed to­600 C overnight. The cans were removed from the freezer for testing, and although some warming mayhave occurred, it should have been small relative to the freezing point of water due to the large bulk offrozen material. The cans are described in Table 1. A I-meter tall column, shown in Fig. 2 (right), wasprepared in a similar manner with ten layers. The first and every other layer were dry (0% ice). Theinterleaving layers were (from top to bottom) 2%, 4%, 6%, 8% and 10% water ice by mass, as shown inFig. 3. It was frozen at -600 C overnight.

9

Figure 2. (Left) Ice/soil can. (Right) I-meter column.

Table 1: SoiVIce Sample Cans.

ID Description Result

Can #1 Three layers of icy regolith, each layer approximately Penetrated all three layers.6 cm thick, 0%, 1%, and 2% water by weight fromtop to bottom.

Can #2 Three layers of icy regolith, each layer approximately Penetrated all three layers, but6 cm thick, 2%, 3%, and 4% water by weight from more slowly than can #1.top to bottom.

Can #3 Three layers of icy regolith, each layer approximately Penetrated the 4% layer and6 cm thick, 4%, 5%, and 6% water by weight from halfway through the 5% layer.top to bottom. Increased frequency did not

renew progress.

Can #4 Three layers of icy regolith, each layer approximately Penetrated the 6% layer and6 cm thick, 6%, 7%, and 8% water by weight from halfway through the 7% layer.top to bottom. Increased frequency did not

renew progress.

Can #5 Pure water ice. Penetrated very quickly.

Can #6 Three layers of icy regolith, each layer approximately Penetrated approximately 56 cm thick, 8%, 9%, and 10% water by weight from cm into the 8% layer.top to bottom.

10

dry

dry

2% Ice

dry

dry

dry

6% Ice

4% Ice

8% Ice

10% Ice

Final rod loc: @5.5cm from edge @5 deg angletoward center (cone & rod progress toward tubeinside sidewall). Rod chuck end sticking 4 emabove top surface. Contact with side wall at 67cmdepth. Data valid only thru 6% water ice layer.

Oem tube top

no regolith---- 8.6 + 1.2cm Top of JSC-la

} 7.7cm thick layer JSC-1a,---- 16.3 + 1.2cm

} 9.6cm thick: 5.5kg JSC-1a+O.llkg H20

---- 25.9 + 1.2cm

} 7.9cm thick layer JSC-1a

---- 33.8 + 1.2cm

} 1O.7cm thick: 5.5kgJSC-1a+O.22kg H20---- 44.5 + 1.2cm

} 9.1cm thick layer JSC-1a---- 53.6 + 1.2cm

} 1O.9cm thick: 5.5kgJSC-1a+O.33kg H20---- 64.5 + 1.2cm

} 8.7em thick layer JSC-1a---- 73.2 + 1.2em

} 10.4cm thick: 5.5kg JSC-1a+o.44kg H20---- 83.6 + 1.2em

} 8.2cm thick layer JSC-1a---- 91.8 + 1.2cm

} Hem thick: 5.5kg JSC-1a+0.55kg H20

l04cm inside -- Bottom of JSC-1a

~p 3.5cm from bottom

---.::L.4cm

rtI_.- ..c."0 uE.::u'-"00. IO_N

Figure 3. One-meter icy/soil column with fmal position of penetrometer.

Procedure

The cone penetrometer was held as shown in Fig. 4 with the cone tip touching the sample can. Onsome occasions, an effort was made to push it into the icy regolith without percussion but with moderatedownforce provided by the operator. For all samples, percussion was activated, providing 2.6 Joules perblow at about a 15 Hz repetition rate. Each can was penetrated using only the 61.7 N weight of thepenetrator for downforce or sometimes using additional downforce provided by the operator pushingdown on the penetrometer's handles. In each case the can was sitting on an Acculab electronic massscale, which provided a measurement of the downforce. Each penetration event, including the reading ofthe mass balance, was video recorded. The rate of penetration was obtained post-test by observing thelength indicators on the shaft of the penetrator as they entered the ice/simulant mixture.

11

Figure 4. Penetration into sample can

Results

Penetration into cans #1 and #2, which had ice contents between 0% and 5% by mass, was notdifficult. The percussive penetrator made progress under its own weight. The speed of penetrationdecreased as the percent water content increased.

In the case of can #3, the percussive penetrator acting under its own weight achieved a maximumdepth of 7.5 cm, passing through the top layer and some portion of the second layer before progresshalted. The top few centimeters of icy regolith was broken into chunks as the cone passed through it.Increasing down force was then applied by the operator. No further progress beyond this was possiblewith only operator applied weight. Video shows that the sheet metal top surface of the mass scale wasvibrating in response to the sample can. It is possible that this motion rendered the percussive cone lesseffective by absorbing some of the percussive energy. To investigate this, the can was moved to the floornext to the scale. The percussive cone was operated with a large down force applied from 2 peopleamounting to about 750 N. There was a small, barely perceptible additional penetration. This was alsotried with the can back on the scale to measure the load with no further cone penetration noted.Penetration into the third layer, 6% ice, was not achieved. Figure 5 shows the depth versus time profilefor can #3, obtained from post-test video analysis. Figure 6 shows the downforce versus time profile.

12

--------------

Percussive Cone 4,5,6% Ice JSC-laa

-1

-2

E~ -3~

~ -4c~ ·5ou

-6

-7

-8

20 40 60 80 10

\.~~ ------ -

o

-ConeDepth

-4%lceJSC-1a

- S%lceJSC-1a

TIme to Depth (sec)

Figure 5. Penetration depth versus time for sample can #3.

Down Force (kg @lG)

::4 + +-.::~~::.::_-_-_-_-_~_c _t---~o 20 40 60 80 100

Figure 6. Downforce versus time for sample can #3.

Can #4 performed similarly to can #3. The cone acting only under its own weight penetrated the firstlayer at 6% ice and a portion of the second layer at 7% ice. Additional down force was appliedmomentarily at ~300 seconds to see if further progress could be made. Minor additional penetration wasachieved. After this the percussor frequency was increased to maximum without significant addedpenetration progress. Video shows that the sheet metal top surface of the mass scale was vibrating inreaction to the sample can. It is possible that this motion rendered the percussive cone less effective byabsorbing some of the percussive energy. Figure 7 shows the penetration depth versus time for can #4,and fig. 8 shows the downforce versus time. Figure 9 shows the fracturing of the sample's surface.

13

Percussive Cone 6,7,8% Ice JSC-la0-.-------,--,-----,-------,--,------,--,---,

-1 .-_~c._&_n-~'n'*l',nl___--1-<>'C'G-lOO 25n '00 350-400

-2 +--------------------

E -3 ~I\<------------------

~ -4 -t-'lr-------------------'5 '-....~ -5 +--""""''''______.=-----------------

~ -6 +----------"".......,-------------u -7 -1--------''''''----===;;;;;:=----­

-8 +----------------.........--=-_=----9 +--------------------

-10 .1.- _

TIme to Depth (sec)

Figure 7. Penetration depth versus time for sample can #4.

-Cone Depth

-6%lceJSC·la

-7%lceJSC-la

Down Force (kg @lG)

:: j ~a so 100 150 200 250 300 350 400

Figure 8. Downforce versus time for sample can #4.

14

-------------------------------------- - --------------------

Figure 9. Fractured surface of can #4. Shaft demonstrates depth of penetration.

In the case of can #5 with 100% water ice (no simulant), the penetrator fractured the ice, and theresulting large pieces moved apart or slid past each other as the cone moved deeper into the target. Thecone penetrated to the bottom of the can much more quickly than it had penetrated the different mixturesthat contained lunar soil simulant. Figure 10 shows the fractured surface of can #5.

Figure 10. Fractured surface of can #5 with 100% water ice.

15

------------

In the case of can #6 with 8% and higher water ice, the cone penetrated the surface but it did not do soby fracturing the top layer of icy regolith into chunks as before. The hole that the cone created in the icyregolith had very clean sides. Powdered material was observed on top of the target's original surface, andthis powdered material could be brushed aside to reveal the original surface. From this, we infer that thecone was pulverizing icy regolith and ejecting it from the resulting borehole though the force of its ownvibration. When the borehole became too deep for the pulverized material to be ejected, progress came toa halt.

For the I-meter column, the cone penetrated its entire length to a depth of 90.7 cm. The rate ofpenetration varied, and was observed to correspond with the layering as shown in Fig. II, where theslopes of the fitted linear segments are the penetration rates. The 8% slope is not valid since the cone wason the container sidewall and the container was splitting, relieving the soil's stress. However, thepenetrator did not maintain a straight vertical path, and struck the side of the column 67.2 cm below thesurface, or 2.6 cm into the 8% layer. We were unable to extract the penetrator from the frozen soil afterdriving it in to its full depth. Figure 12 shows the top surface (unfractured) with the rod still embedded inthe icy regolith after the percussive penetrometer was de-attached.

-1 -Cone Penetration

--Layer-1 (Drf JSC-1a)y = -U.3~ IIx - lS.~ 114

-Layer-2 (2%lce JSC-1a)

-I <'lypr-~ (Dry JSC-1 <1)

-Layer-4(4%lce JSC-Ia)y= -O.1343x-19.434

-Layer-5(Drf JSC-Ia)IIIu

.::::.. -Layer-6 (6%lce JSC-1a)III....v --Layer-7(Dry JSC-1a)III-.~ --Layer-8 (8%lce JSC-1a)E y = -0.0463x - 38.04u -Layer-9(Drf JSC-Ia)~- -Layer-IO (lO%lce JSC-la)CoIII0III -Down Load Kg@ IGc0u -4%lceSlope

--6%lceS ope

--8%lccS ope

-2%lceSope

--Linear (4%lceSlope)

--Linear (6%lceSlope)

--Linear (8%lceSlope)Time to Depth (Seconds)

Figure 11. Depth of penetration and downforce versus time for I-meter column.

16

Figure 12. Top surface of I-meter column with embedded penetrometer rod.

Discussion

Ice Content

In general we find as expected that, in the range of 0% to 10% ice content by mass, the greater icecontent is more resistant to penetration. However, it is interesting that the top layer of can #4 with 6% icecould be penetrated, whereas the second layer of can #3 with only 5% ice could not be penetrated. Thus,we found that it is not simply the percentage of ice that matters, but whether the fracturing ice has room tomove into the surrounding volume. In the case of top layers, there was always room at the free surfacefor fractured chunks of ice to move upward. In the second and deeper layers of the cans, the ice could notalways fracture and move. Thus, a lower percentage of ice at depth could resist penetration whereas ahigher percentage of ice at the surface could not.

The mechanics are apparently different than the case of penetration into dry regolith. Dry regolith isfree to rearrange at the grain-scale to make room for the penetrator. The strain field of cone penetrationinto ordinary, terrestrial sands and soils has been studied in detail (see for example [Tumay at aI, 1985]and [Acar and Tumay, 1986]), indicating soil motion (of decreasing amplitude) at long distances from thecone. Furthermore, comminution of the individual grains allows their material to move into the porespaces between other grains, increasing the bulk density of the material around the penetrometer to makethe room. For frozen soils, however, the grains cannot move individually, and comminuted material maynot be able to move into pore spaces between neighboring grains. Therefore, to make room for a coneand rod, the frozen soil must exhibit a combination of pulverization with powder removal and fracturingwith relative motion of the fractured domains. As long as the cone is near the surface of the icy regolith,powder could exit the downshaft around the sides of the rod, and likewise near the surface the fractureddomains could move upward above the free surface of the sample.

17

-- ---------

For the small sample cans with three layers of icy/soil mixture, when the ice content was sufficientlyhigh the cone could not penetrate through the second layer because it was trapped between the overlyingand underlying frozen layers. For sample cans #1 and #2 this was not a problem. Presumably the low icecontent did not stop the soil from deforming at the grain scale to densify within each layer, or to push intothe neighboring layers that then deformed at the grain scale to absorb the additional volume of material.Therefore, it appears that the transition in penetration mechanics occurs somewhere in the range of 3% to5% ice content by mass.

For sample can #5 with 100% water ice, the soil did fracture but was able to rearrange all the way tothe bottom of the can, permitting the penetrator to reach the bottom. Apparently, the ice has less frictionthan an ice/regolith mixture, and thus the fractured chunks even deep in the can are able to push thefractured chunks above them out ofthe way.

For the case of the I-meter column, the fractured domains could make room by expanding into theinterleaving dry layers of regolith. A fractured domain expanding into neighboring space wouldpresumably experience more resistance if it were moving into dry regolith than if it were moving intoempty space above the free surface of the sample, as was the case with the top layers of the smallersample cans. However, the dry regolith did not produce enough resistance to stop this penetration asevidenced by the cone passing successfully through layers of up to 10% ice. To make room for adjacentfractured domains, the dry regolith layers must have densified through the ordinary processes discussedabove for cone penetration. However, in this experiment only half the volume of the column was dryregolith whereas the other half was icy. Half the domain must have been adequate to absorb the fullvolume of the penetrating cone and rod.

Based on these results, it is likely that bulk volumes of greater than 4-6% ice (by mass) will beresistant to percussive cone penetrometers unless a method is developed to remove pulverized material.Fortunately, the instrument that will go onto the cone, which is being developed in the companion project,is capable of detecting and measuring ice content at concentrations less than these values. Also, since thepercussive penetrometer can always penetrate at least the top few centimeters of material higher than 4­6% (since the fractures can expand upward into the material that has less ice), the system is guaranteed toenter at least that concentration of ice even if no modifications are made. If the lunar ice lies beneath ameter of desiccated soil, or soil that contains less than 4-6% ice, or layers less than a few centimetersthick of any concentration interleaved by layers less than 4-6% ice, then the system will successfully passthrough it to the bulk quantities of more concentrated ice. In any case, when it finds the bulk of the moreconcentrated ice, it should penetrate several centimeters. Therefore, the delivery method appears to besuccessful but with room for increased performance.

Penetration Rate

The penetration rate was not a strong function of downforce. In most cases it was a strong functiononly of the ice content. Therefore, penetration rate may serve as a useful secondary measurement of icecontent to corroborate what is measured by the primary instrument. By comparing the primaryinstrument's [mdings with the penetration rate, it may also be possible to back out information about thecompaction of the soil and thus the volume of pore space not occupied by frozen volatiles. This maysupport analysis of the permeability of the regolith and modeling of ice stability and transportmechanisms.

It makes sense that penetration rate would be dependent on ice content. The ice must be fractured orpulverized to permit movement of material to admit the cone. The more volume ice there is bonding thesoil grains together, the more energy that must be expended to break those grains apart. Thus, morepercussive blows are needed to free equivalent volumes of regolith when there is more ice. Penetrationcould be sped up by increasing the percussion rate, but (as seen with can #4) if the blows are incapable ofmoving material, then increasing their frequency will not restore motion.

18

On the other hand, toward the bottom of the I-meter column, increased downforce did helppenetration (see "Download" curve on Fig. 11). This might be because the fracturing ice layers deep inthe column needed to be mechanically pushed into the neighboring volumes of dry regolith. Because thelarge chunks would be moving dry regolith far away from the percussing cone, their motion relied uponthe direct downforce. Thus, there is additional information available by analyzing both downforce andpenetration rate together that may indicate the structure of ice layers beneath the surface.

Retracting the Rod

In the one test with the I-meter column, the rod could not be retracted. This may have been due tothe fact that the rod became bent as it struck the wall of the column, but other factors may havecontributed. For example, the base diameter of the cone is larger than the diameter of the rod, andcompacted JSC-I a can exert high friction on a rod that has been driven into it. There is also someconcern that ice could freeze to the cone or shaft. The instrument on the cone will have a heater elementto enable volatilization of the ice as a part of measuring its concentration. The heater element could servea secondary function of de-icing the regolith around the cone or shaft to help free it. Also, the team hasindentified mechanical changes to the design ofthe system to enable easier retraction.

Summary of Findings1. A percussive cone can penetrate icy regolith at ice concentration layers that a static cone cannot

penetrate. The percussive penetrator was able to penetrate material under 65 N of downforcethat the static cone could not penetrate under full body weight.

2. The percussive cone could penetrate:a. 100% water ice (-60 C);b. dry soil that is compacted and cold (-60 C);c. ice/soil mixtures up to 4-6% ice by weight (note that 5% ice is more resistant than 100%

ice) with much less resistance than non-percussive;d. mixtures with 6% and 8% ice as long as there is a free surface above the layer to allow

the icy chips room to move;e. a little more than a cone-length into the top surface of mixtures that have 8% or more ice;f. completely through layers of 8% ice as long as they are interleaved with dry layers to

allow the icy chips room to move.

3. These percentages of ice are within the range that can be detected by the sensor developed in thecompanion project, "ISDS for Water Detection on the Lunar Surface." Therefore, the systemis capable of penetrating deeply enough into the regolith to detect ice.

4. The ability of a percussive cone to displace material affects its ability to penetrate material. Thedevice proved capable of penetrating material with 8% ice, but did not penetrate very farbecause it could not displace the resulting chip. A certain amount of material must be displacedfor the cone to advance.

5. Increased downforce on the percussive system did not result in increased penetration capability.In hard material, the percussive penetrator made no more progress under 750 N than it didunder 65 N. This suggests that increasing the energy delivered in each percussive blow wouldbe a more effective than increasing downforce for penetrating stronger materials. When the ice

19

is too dense, pushing harder will not make it penetrate. The percussive system eitherpenetrates or doesn't.

6. There may be cases with layers of ice interleaved by dry regolith in which increased downforcehelps move the fractured ice and thus increases penetration rate, but this condition is not thebaseline expectation for lunar regolith. If such a condition does exist on the Moon, then it canbe detected by measuring both penetration rate and downforce to corroborate other instrumentson the cone.

7. A percussive cone can become stuck in frozen regolith. For anchoring, this is beneficial. A roddriven in under 65N (15 Ibs) of downforce could not be pulled out with at least that muchforce. For repeated probing, this must be addressed.

Conclusions

This investigation successfully demonstrated percussive penetration of regolith with varying iceconcentrations. It demonstrated that percussive penetration is feasible in situations where non-percussiveis not. It successfully demonstrated penetration to a meter in depth, which is the expected depth to lunarice. It demonstrated that the device is capable of entering ice concentrations that are easily within themeasurement range of instruments that will go on the cone. This serves as experimental proof-of-conceptof the critical function, and so percussive insertion of lunar ice instruments has now achieved TechnologyReadiness Level 3.

The experiments developed preliminary correlation data between penetration rate and ice content, andthus the data from the penetration process can be used to corroborate the findings of another sensor. Thisproject also produced insights into the mechanics ofthe penetration resistance of ice and ice/soil mixtures.It provided an opportunity to develop and test methods to prepare ice/soil mixtures for mechanical testing.It provided insights into how to improve the test stands and hardware. It also indicated a number ofmodifications that may be made to improve the penetration system, which will be the subject of on-goingprojects.

20

ReferencesAcar, Yalcin B., and Mehmet T. Tumay (1986). "Strain Field around Cones in Steady Penetration," J.

Geotech Eng 112 (2), 207 - 13.

Alshibli, Khalid A., and Alsidqi Hasan (2009), "Strength Properties of JSC-IA Lunar Regolith Simulant,"J. Geotech. and Geoenvir. Engrg. 135 (5),673-679.

Carrier, W. David, III, Gary R. Olhoeft and Wendell Mendell, "Physical Properties of the Lunar Surface,"in Lunar Sourcebook, A User's Guide to the Moon, G. H. Heiken, D.T. Vaniman and B.M. French,eds., (Cambridge University Press, Melbourne, Australia, 1991), pp. 475 - 594.

Chen, K., J. Cole, C. Conger, J. Draskovic, M. Lohr, K. Klein, T. Scheidemantel and P. Schiffer(2006), "Granular materials: Packing grains by thermal cycling," Nature 442, 257.

Gamsky, Jacob N., and Philip T. Metzger (2010), "The Physical State of Lunar Soil in the PermanentlyShadowed Craters of the Moon," paper presented at Earth and Space 2010, 12th Biennial ASCEAerospace Division International Conference on Engineering, Construction and Operations inChallenging Environments, Honolulu, HI, Mar. 14-17,2010.

Gertsch, Leslie, Robert Gustafson, Richard Gertsch (2006), "Effect of water ice content on excavatabilityof lunar regolith," paper presented at Space Technology and Applications International Forum-STAIF 2006, AlP Conference Proceedings, vol. 813 (AlP, Melville, NY), pp. 1093-1100.

Klosky, J.L., S. Sture, H.-Y. Ko, and F. Barnes (2000), "Geotechnical behavior of JSC-l lunar soilsimulant," J. Aerospace Eng. 13 (4), 133-138.

Tumay, Mehmet T., Yalcin B. Acar, Murat H. Cekirge, and Narayanan Ramesh (1985), "Flow FieldAround Cones in Steady Penetration," J. Geotech Eng 111 (2), 193 - 204.

Zeng, Xiangwu, Chunmei He, Heather Oravec, Allen Wilkinson, Juan Agui, and Vivake Asnani (2010),"Geotechnical Properties ofJSC-l A Lunar Soil Simulant," J. Aerosp. Engrg. 23 (2), 111-116.

21

REPORT DOCUMENTATION PAGE1

Form ApprovedOMS No. 0704-0188

The public reporting burden for this collection of information is estimated to average I hour per response, including the time for reviewing instructions, searchingexisting data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding thisburden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Department of Defense, WashingtonHeadquarters Services, Directorate for Information Operations and Reports (0704-0188),1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302.Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection ofinformation if it does not display a currently valid OMB control number.PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS.1. REPORT DATE (DD-MM-YYYY) 12. REPORT TYPE 13. DATES COVERED (From- To)

Final June 2010 - Sept. 20104. TITLE AND SUBTITLE Sa. CONTRACT NUMBER

Low Force Icy Regolith Penetration Technology N/A5b. GRANT NUMBER

5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S)5d. PROJECT NUMBER

P. T. Metzger, G. M. Galloway, and J. G. Mantovani (1)5e. TASK NUMBER

K. Zacny and J. Craft (2)Sf. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION

(1) NASA Kennedy Space Center, FL 32899REPORT NUMBER

(2) Honeybee Robotics, Inc., 460 W. 34th Street, New York City, NewYork 10001

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSORINGIMONITOR'S ACRONYM(S)

NASA Innovative Partnerships Program

11. SPONSORING/MONITORINGREPORT NUMBER

12. DISTRIBUTION/AVAILABILITY STATEMENT

Unclassified - Unlimited Distribution: StandardSubject Category:Availability: Publicly Available STI

13. SUPPLEMENTARY NOTES

N/A

14. ABSTRACT Recent data from the Moon, including LCROSS data, indicate large quantities of water ice and other volatilesfrozen into the soil in the permanently shadowed craters near the poles. If verified and exploited, these volatiles willrevolutionize spaceflight as an inexpensive source of propellants and other consumables outside Earth's gravity well. Thisreport discusses a preliminary investigation of a method to insert a sensor through such a soiVice mixture to verify thepresence, nature, and concentration of the ice. It uses percussion to deliver mechanical energy into the frozen mixture,breaking up the ice and decompacting the soil so that only low reaction forces are required from a rover or spacecraft to pushthe sensor downward. The tests demonstrate that this method may be ideal for a small platform in lunar gravity. However,there are some cases where the system may not be able to penetrate the icy soil, and there is some risk ofthe sensor becomingstuck so that it cannot be retracted, so further work is needed. A companion project (ISDS for Water Detection on the LunarSurface) has performed preliminary investigation of a dielectric/thermal sensor for use with this system.

15. SUBJECT TERMS

17. LIMITATION OF 18. NUMBER 19a. NAME OF RESPONSIBLE PERSON16. SECURITY CLASSIFICATION OF: ABSTRACT OF

UU PAGESa. REPORT Ib. ABS~RACT Ic. THI~AGE

U19b. TELEPHONE NUMBER (Include area code)

Standard Form 298 (Rev. 8-98)Prescribed by ANSI Std. Z39